may have been developed.
Miohippus, Protohippus, and Hypohippus, varying in size from that of a
sheep to that of a donkey, are other branches of this spreading family.

In the Pliocene period the evolution of the main stem culminates in
the appearance of the horse, and the collateral branches are destroyed.
Pliohippus is a further intermediate form. It has only one toe on each
foot, with two large splint bones, but its hoof is less round than that
of the horse, and it differs in the shape of the skull and the length
of the teeth. The true horse (Equus) at length appears, in Europe and
America, before the close of the Tertiary period. As is well known, it
still has the rudimentary traces of its second and fourth toes in the
shape of splint bones, and these bones are not only more definitely
toe-shaped in the foal before birth, but are occasionally developed and
give us a three-toed horse.

From these successive remains we can confidently picture the evolution,
during two or three million years, of one of our most familiar mammals.
It must not, of course, be supposed that these fossil remains all
represent "ancestors of the horse." In some cases they may very well
do so; in others, as we saw, they represent sidebranches of the family
which have become extinct. But even such successive forms as the
Eohippus, Mesohippus, Miohippus, and Pliohippus must not be arranged
in a direct line as the pedigree of the horse. The family became most
extensive in the Miocene, and we must regard the casual fossil specimens
we have discovered as illustrations of the various phases in the
development of the horse from the primitive Ungulate. When we recollect
what we saw in an earlier chapter about the evolution of grassy plains
and the successive rises of the land during the Tertiary period, and
when we reflect on the simultaneous advance of the carnivores, we can
without difficulty realise this evolution of our familiar companion from
a hyrax-like little animal of two million years ago.

We have not in many cases so rich a collection of intermediate forms as
in the case of the horse, but our fossil mammals are numerous enough
to suggest a similar development of all the mammals of to-day. The
primitive family which gave birth to the horse also gave us, as we saw,
the tapir and the rhinoceros. We find ancestral tapirs in Europe and
America during the Tertiary period, but the later cold has driven them
to the warm swamps of Brazil and Malay